Q&A: How AI can empower creativity

The AI ecosystem really is accelerating at a great rate of change, and we are seeing new applications developed every day, from life sciences, healthcare, to cyber security. It’s going to impact every single industry that we know. That offers a great promise to us, I think, of AI really changing and helping to augment the way that we live and work in a positive way.

There are two things for me. One, access to the infrastructure, and two, regulation. 

It’s become very clear that access to the GPUs on which large language models (LLMs) are trained is limited, including the space in data centres required to house the GPUs.

Secondly, and importantly, regulation requires partnership from subject matter experts and industry leaders to ensure that it isn’t limiting innovation in a widespread way. So you see things like the European AI Act, which is a good starting point for a lot of this, but really needs to be refined with the help of subject matter expertise. 

As industry leaders, we really have an accountability to work with government to shape that in the right way when it comes to infrastructure, such as access to the GPUs and the data centres. There’s also rising demand on power grids, which we are seeing take a toll now. 

But all of these things are kind of ancillary to making sure you have data and you have the talent. There is a real need to educate people and make sure that they are really focusing on getting access to the right things, using the data in the right way and developing the skills that will continue to advance AI. 

The answer really is democratisation.

At Northern Data Group, for instance, we are very much focused on providing access to startups and other smaller industries to ensure that they can work with this technology in a way that allows them to accelerate and commoditise their applications. 

And when I say we democratise access, what we do is we run what is called an AI accelerator. That means we invite startups to come and use our infrastructure at no cost to them. And we also then support them. 

So back to my points around education and using people in the right way, alongside the technology, we actually help to mentor them, guide them and give them access to industry expertise. We use partners like HP, Gigabyte and NVIDIA in this case to really help these startups to fully understand what they are training and how to bring that to life very quickly. 

There are a lot of different places to use it and I think ChatGPT obviously gets a lot of focus because it’s a very useful application. But of course there are applications like I mentioned in healthcare, life sciences, but also in the creative space, where I think we have seen some very, very nice applications of the intersection between human capability and the use of this technology.

We’ve actually been involved in a really special, groundbreaking effort with Shona Heath, who’s the award winning, or actually I should say Oscar winning, production designer for Poor Things. 

We worked with Shona to bring to life that intersection between human capability and technology, and Shona worked with us to use AI in one of her films. We were really amazed to see this beautiful representation of when AI is used as a co-intelligence to augment human capability. 

We were able to see the film that Shona created both as the analogue version as well as the all-AI version, and then the marriage of those two outcomes in a third version, which was, without a doubt, one of the most beautiful things to see. It really was a demonstration of how, when directed in the right way and when used in the right way, AI can be something that helps us to make something even more beautiful.

I think it really will help us to augment capabilities, automate things and really elevate some of the things that we do creatively.

But I also think that there is an emerging and controversial question around the protection of that creative work. And this is where there’s been a lot of controversy because we need to find ways to protect data and protect intellectual property in a way that artists feel safe to use this type of technology.

There certainly is a discussion to be had around that, but I think this is a game where companies like Northern Data can be helpful because ultimately we don’t operate in the same way that a hyperscaler would, for instance.

So when you are training your LLM to support your creative efforts, we offer data sovereignty. And I think that this is something that the industry needs to be very much more aware of. Businesses need to ensure that they are using and working with providers that actually help to protect their data.